Large-Scale Comparison of Univariate and Multivariate Meta-Analysis for Categorical Outcomes
Thomas Trikalinos, Brown University
Treatment effects for multiple outcomes can be meta-analyzed separately or jointly, but no systematic empirical comparison of the two approaches exists. Among the more than 5,000 reviews in the Cochrane Library of Systematic Reviews, we identified 45 reviews, including 1,473 trials and 258,675 patients, that contained two or three univariate meta-analyses of categorical outcomes for the same interventions that could also be analyzed jointly. Eligible were meta-analyses with at least seven trials reporting all outcomes for which the cross-classification tables were exactly recoverable (e.g., outcomes were mutually exclusive or one was a subset of the other). We describe comparisons between the methods for main effects and for functionals of main effects estimated with univariate and multivariate models based on discrete and approximate likelihoods.